Learning Device, Defect Detection Device, And Defect Detection Method

Abstract: Provided is a learning device comprising: a training time-series data acquisition unit (101A) that collects training time-series data, which is acquired by the same or the same type of target device as a monitoring target device or by a sensor installed near the target device, and configuration parameter data of the target device or environment data on the target device while associating same with each other; a segment set generation unit (102) that divides the training time-series data into training segments, which are each partial time-series data indicating an operation state including both of a rise from a first value to a second value and a fall from the second value to the first value in a waveform represented by the training time-series data, and generates a segment set including the plurality of training segments; a segment set sorting unit (103) that uses the configuration parameter data or the environment data to group the plurality of training segments included in the generated segment set into similar training segments and classify the similar training segments into at least one similar segment set; and a sample segment generation unit (104) that generates a sample segment indicating a normal range of the operation of the target device from a plurality of training segments included in the at least one similar segment set.

Patent Information

Applicants

Inventors

Specification

Documents

Application Documents

Search Strategy